Mismatch between subjective and objective dysautonomia

In dysautonomia evaluation, symptom questionnaires and clinical symptom reports are frequently used as proxies for physiological impairment. The assumption is that patients who report more severe symptoms have more severe autonomic dysfunction, and patients who appear to be managing have less severe dysfunction. A 2024 paper by Novak and colleagues from Brigham and Women's Faulkner Hospital's Autonomic Laboratory tests this assumption at scale — 2,627 patients, comprehensive symptom questionnaires, and standardized objective autonomic testing — and finds that it does not hold. The correlation between subjective symptom burden and objective autonomic impairment is essentially zero. Symptom reports and objective measurements are measuring different things, and neither can substitute for the other. The practical consequence is that autonomic evaluation requires objective testing regardless of how the patient appears to be functioning, and the severity of patient presentation is not a reliable guide to what the testing will show.

The Study: Scale and Comprehensiveness

The Novak 2024 paper draws on the Brigham and Women's Faulkner Autonomic Laboratory's clinical database, analyzing data from 2,627 patients who completed both validated autonomic symptom questionnaires and a standardized battery of objective autonomic tests. The scale is notable — 2,627 patients is large for an autonomic research study and provides adequate statistical power to detect even modest correlations. If a real correlation between symptom burden and objective impairment existed in this population, this study had sufficient power to find it.

The symptom questionnaires used were the Survey of Autonomic Symptoms (SAS) and the Composite Autonomic Symptom Score 31 (COMPASS-31). Both are validated instruments designed to quantify the severity of autonomic symptom burden across multiple symptom domains. The objective testing battery included tilt table testing, Valsalva maneuver, deep breathing assessment, sudomotor function testing, end-tidal CO₂ monitoring, and transcranial Doppler cerebral blood flow measurement. The combination of heart rate variability analysis, blood pressure responses, sweat gland function, capnography, and cerebral blood flow provides a multi-domain objective picture of autonomic function that goes significantly beyond what standard clinical evaluations assess.

The diagnostic composition of the patient sample was heterogeneous: POTS, ME/CFS, MCAS, Ehlers-Danlos syndrome, post-COVID dysautonomia, Lyme disease, and other autonomic conditions. This diagnostic breadth is important because it tests the symptom-to-objective correlation across the full spectrum of conditions that autonomic labs evaluate, not in a single-disease cohort where the relationship might be artificially strong.

The Core Finding: Symptom Burden and Objective Impairment Do Not Correlate

The internal consistency of the measurement instruments was strong. The two symptom questionnaires correlated well with each other — patients reporting high symptom burden on COMPASS-31 tended to report high symptom burden on SAS. This confirms that the questionnaires are reliably measuring subjective symptom experience. The objective test measures also correlated well internally across testing modalities — patients showing more impaired tilt responses tended to show more impaired Valsalva and HRV results. This confirms that the objective battery is reliably measuring physiological impairment.

But the cross-domain correlation — subjective symptom burden against objective impairment severity — was essentially zero. Patients who scored in the most severe range on symptom questionnaires showed wide variation in objective testing outcomes, from minimally impaired to severely impaired. Patients who scored in the mild range on symptom questionnaires similarly showed wide variation in objective outcomes, including some patients with severely impaired objective results despite reporting relatively mild symptom burden.

The mismatch appeared in both directions and was consistent across every diagnostic subgroup in the study. It was not a feature of POTS patients specifically, or ME/CFS patients specifically, or post-COVID patients specifically. It was a systematic pattern across the entire heterogeneous autonomic patient population. The disconnect between symptom report and objective measurement is not a quirk of any one condition. It appears to be a general feature of how autonomic dysfunction presents.

Two Directions of Mismatch: What Each Means

The mismatch between symptoms and objective findings runs in two clinically important directions, and each deserves separate attention.

The first direction: patients with severe objective autonomic impairment who report relatively mild symptom burden. This is the pattern that creates the most clinical risk if symptom report guides evaluation depth. A patient who appears to be managing reasonably well may have severely abnormal autonomic function on objective testing. The normal clinical assumption — that patients who appear functional have adequate autonomic regulation — is not supported by the Novak data. The objective abnormality may be present without generating the symptom experience that would flag it as severe. This happens through the interoceptive adaptation mechanisms described in the Freeman orthostatic hypotension research: chronic dysregulation recalibrates the baseline against which deviation is perceived as symptomatic. A patient who has functioned with impaired autonomic regulation for years may not experience that regulation as severely abnormal anymore, because it is their normal.

The second direction: patients with high subjective symptom burden who show minimal objective autonomic impairment. Novak and colleagues identify this subgroup as potentially having what they term "idiosyncratic autonomic-like symptomatology" — a presentation in which the symptom experience resembles dysautonomia but the objective autonomic physiology is not severely impaired. This subgroup may require different clinical evaluation and management than patients with structural autonomic dysfunction, because the mechanism producing their symptoms may not be primarily the autonomic nervous system itself.

Both directions have diagnostic and therapeutic implications. Neither is captured by relying on symptom report alone.

Why Questionnaires Are a Starting Point, Not a Destination

Validated autonomic symptom questionnaires like COMPASS-31 are useful clinical tools. They systematically capture the symptom domains that are relevant to autonomic evaluation and provide a standardized way to compare symptom burden across patients and across time within the same patient. They are appropriate as screening tools, as outcome tracking instruments, and as part of a comprehensive clinical picture.

What the Novak data establish is that they cannot substitute for objective testing as a measure of autonomic physiological impairment. The poor correlation between questionnaire scores and objective results means that you cannot use questionnaire results to predict what objective testing will show — in either direction. A patient with a high COMPASS-31 score needs objective testing to determine whether that score reflects severe physiological impairment or a high symptom-to-physiology ratio. A patient with a low COMPASS-31 score cannot be assumed to have normal objective physiology.

The clinical model that this paper argues for is one in which objective autonomic testing is the primary assessment of physiological impairment, and symptom questionnaires are a complementary but non-substitutable measure of symptom experience. The two data streams provide different, non-interchangeable information. Combining them gives a fuller picture. Using either alone to draw conclusions about the other produces clinical errors.

What This Means for Patients Navigating Evaluation

For patients in evaluation, the Novak finding provides a specific framing for clinical conversations. Appearing to manage — not presenting as acutely distressed, maintaining a functional facade during a clinical appointment — is not evidence of normal autonomic physiology. The paper establishes directly that patients with severely impaired objective autonomic testing sometimes report mild symptoms. If your evaluation relies on how you present or how your symptoms score on a questionnaire without including objective physiological testing, it is not measuring your autonomic function. It is measuring your symptom experience, which may or may not reflect what your autonomic system is doing.

The objective testing battery in the Novak study — tilt, Valsalva, deep breathing, sudomotor, CO₂, and TCD — represents a comprehensive autonomic evaluation that goes beyond what most clinical settings provide. For patients who have been evaluated only with symptom questionnaires and basic vital signs, the Novak data establish that the component of evaluation most relevant to identifying physiological impairment has not been performed. How you feel is real and clinically relevant. But it is not a substitute for measuring what your physiology is doing.